Iron-mediated regulation of liver hepcidin expression in rats and mice is abolished by alcohol

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Alcohol reduces and iron increases liver hepcidin synthesis. This study investigates the interaction of alcohol and iron in the regulation of hepcidin messenger RNA (mRNA) expression in animal models. Mice were administered 10% ethanol for 7 days after an iron-overloaded diet. Rats were administered regular or ethanol-Lieber De Carli diets for 7 weeks with or without carbonyl iron. Hfe−/− mice were used as a model for genetic iron overload. Hepcidin mRNA expression was determined by real-time polymerase chain reaction (PCR) and northern blotting. Iron elevated and alcohol decreased liver hepcidin expression in mice and rats. Interestingly, despite iron overload, alcohol was capable of suppressing the up-regulation of hepcidin mRNA expression in both models. Liver iron and ferritin protein expression was elevated in alcohol-treated rats, but was not elevated further in rats treated with both iron and alcohol. Duodenal ferroportin protein expression was increased both in alcohol-treated mice and in mice treated with alcohol and iron. Hfe−/− mice treated with ethanol for 7 days exhibited a further decrease in hepcidin mRNA expression. The iron-induced increase in DNA-binding activity of the transcription factor CCAAT/enhancer binding protein alpha (C/EBP alpha) was also suppressed by alcohol. Conclusion: Alcohol abolishes the iron-induced up-regulation of both liver hepcidin transcription and the DNA-binding activity of C/EBP alpha. Of note, hepcidin protects the body from the harmful effects of iron overload. Our findings therefore suggest that alcohol negates the protective effect of hepcidin, which may have implications for the liver injury observed in alcoholic liver disease and genetic hemochromatosis in combination with alcohol. (HEPATOLOGY 2007.)

Hepcidin is a circulatory peptide that is synthesized in the liver and acts as the key regulator of iron metabolism by modulating iron absorption through the duodenum and the release of iron from macrophages.1, 2 We have recently demonstrated that acute alcohol-induced oxidative stress down-regulates liver hepcidin transcripts in mice by altering the DNA-binding activity of the transcription factor CCAAT/enhancer binding protein (C/EBP) alpha.3 Alcohol-mediated down-regulation of liver hepcidin mRNA expression occurred without associated changes in liver histology or triglycerides.3 Mice exposed to ethanol for 7 days exhibited elevated iron transporter protein expression in the duodenum. This was abolished by injecting mice with hepcidin peptide.3 Bridle et al. also have reported reduced hepcidin mRNA expression in rats with chronic alcohol exposure.4 These results demonstrated a role for alcohol in the regulation of iron metabolism by regulating the expression of hepcidin in the liver.

Liver hepcidin messenger RNA (mRNA) expression is regulated inversely by iron and alcohol.2–4 Namely, iron up-regulates, whereas alcohol down-regulates, liver hepcidin expression. However, the combined effect of alcohol and iron in the regulation of hepcidin expression is unknown. Despite increased body iron stores, patients with genetic hemochromatosis and Hfe-deficient mice express reduced liver hepcidin expression.5, 6 Alcohol consumption increases the risk for liver injury in genetic hemochromatosis patients.7 However, the underlying molecular mechanisms are unclear. Iron has been reported to activate nuclear factor kappa B and elevate tumor necrosis factor alpha–mediated liver injury in an experimental animal model of alcoholic liver disease.8

Hepcidin plays a central role in the regulation of iron homeostasis.2 Hence, understanding the interaction of alcohol and iron in the regulation of hepcidin expression is clinically relevant. It may help clarify our understanding of mechanisms underlying the liver injury caused by iron and alcohol in diseases such as alcoholic liver disease and genetic hemochromatosis in combination with alcohol intake. These studies using animal models of alcohol exposure demonstrate that alcohol abrogates the protective effect of hepcidin in iron overload by rendering the regulation of hepcidin mRNA expression in the liver insensitive to iron.

Abbreviations

c/EBP, CCAAT/enhancer binding protein alpha; Hfe; genetic hemochromatosis protein; mRNA, messenger RNA; PCR, polymerase chain reaction.

Materials and Methods

Animal Experiments

Animal experiments were approved by the animal ethics committee at the University of Nebraska Medical Center. The rodent chow diet used as regular food by the animal facility for rats (Harland Teklad 8656) and mice (Harland Teklad 7012), contained 346.64 mg/ kg and 284.11 mg/ kg iron, respectively. Of note, hepcidin is regulated by iron, and iron overload up-regulates liver hepcidin expression.2 Thus, to achieve a basal hepcidin expression level in all of the animals before the start of each experiment, they were fed initially with a custom-prepared egg-white–based solid rodent diet9 containing 0.02 % carbonyl iron (F614, Bio-Serv, Inc.) for 1 week.

Mouse Models and Experiments.

129/Sv strain wild-type male mice (Charles River Laboratories) were fed with a solid rodent diet (see above) containing 2% carbonyl iron (F616, Bio-Serv, Inc.) for 3 weeks, to create an iron-overload state. All mice consumed equal amounts of diet and did not exhibit any obvious weight differences (25 ± 2 g average weight). Transgenic mice, homozygous for the null allele of Hfe (Hfe−/−) on 129/Sv genetic background, were provided by Dr. Nancy Andrews (Harvard Medical School, Boston) and were created as described previously.10 Hfe−/− mice were fed with the regular chow diet. For alcohol treatment(s), all mice were housed in individual cages and exposed to 10% ethanol in the drinking water or plain water (control) for 7 days, as described previously.3

Rat Models and Experiments.

Wistar male rats (Charles River Laboratories), maintained routinely on rodent chow diet, were switched to the custom-prepared diet containing 0.02% carbonyl iron, as described previously. Subsequently, the rats were housed individually in metabolic cages and pair-fed with either regular or ethanol-containing Lieber De Carli liquids diets (Dyets, Inc., cat no: 710027, 710260, respectively), as described.11 The ethanol content of the diet was gradually increased over an 8-day period (no ethanol for day 1, one-fourth the amount for days 2 to 3, and half the amount for days 4 through 7) to the full amount (36% of total calories as ethanol). Rats were exposed to the full amount of ethanol for 7 weeks. For experiments involving both iron and ethanol, custom-prepared regular or ethanol-Lieber De Carli liquid diets, containing 2% (wt/vol) carbonyl iron (Dyets, Inc., cat no: 710331, 713602, respectively), were diluted 1:3 (vol/vol) with the relevant liquid diet containing no carbonyl iron.

RNA Isolation and Northern Blotting

RNA isolation and northern blotting were performed as described previously.3 Hybridization was performed with 32P-labeled mouse hepcidin, and mouse beta-actin DNA probes random labeled by a commercial kit (Amersham, RPN 1633). Cloning of mouse hepcidin was as previously described.3 Three hundred base pair mouse hepcidin DNA fragment was obtained by Not1 digestion and gel purification; 1,076 bp mouse beta-actin DNA was obtained commercially (Ambion Inc.).

Complementary DNA Synthesis and Real-Time Quantitative Polymerase Chain Reaction Analysis

Complementary DNA synthesis and quantitative polymerase chain reaction (PCR) were performed, as published previously.3 The sequence of Taqman fluorescent probes (5′ 6-[FAM]; 3′ [TAMRA-Q]) and primers are shown in Table 1.

Table 1. Real-Time Quantitative PCR Probe and Primer Sequences
GeneForward Primer (5′–3′)Reverse Primer (5′–3′)Taqman Probe
Mouse hepcidin 1TGCAGAAGAGAAGGAAGAGAGACACACACTGGGAATTGTTACAGCATTCAACTTCCCCATCTGCATCTTCTGCTGT
Rat hepcidinTGAGCAGCGGTGCCTATCTCCATGCCAAGGCTGCAGCGGCAACAGACGAGACAGACTACGGC

Antibodies and Western Blotting

Anti-ferroportin antibody was created by injecting a rabbit with a keyhole limpet–conjugated 23–amino acid peptide corresponding to amino acids 159 through 181 and with a bovine serum albumin–conjugated 22–amino acid peptide, corresponding to amino acids 239 through 260 of rat ferroportin protein (NCBI Entrez protein accession number AAK77858). The unconjugated form of these peptides was employed for competition studies to characterize the specificity of the antibody. Ferritin heavy chain (H-ferritin) antibody was purchased from Santa Cruz Biotechnology Inc. (sc-14416; 1:500 dilution). Tissue lysates and western blots were performed as described previously.3

Electrophoretic Mobility Gelshift Assay

The binding activity of C/EBP was examined by electrophoretic mobility gelshift assay, as described.3, 12 C/EBPα overexpressed in HEK (human embryonic kidney) 293 cells was used as a positive control. pAdTrackC/EBPa plasmid13 was transfected into HEK293 cells using Fugene 6 (Roche Diagnostics). Nuclear extracts were isolated 16 hours after the transfections, and 2 μg nuclear extract protein was used for the gelshift assays.

Liver Iron

Hepatic iron levels were quantified by commercial plasma–atomic emission spectrometry (Covance, Inc.), as described.14

Statistical Analysis

Statistical analysis of differences in the various treatment groups was performed using analysis of variance.

Results

To study the combined effect of alcohol and iron in the regulation of hepcidin mRNA expression, we employed animal models of alcohol exposure, as described in Materials and Methods. Rats fed with ethanol-containing Lieber De Carli liquid diet displayed down-regulation (0.5 ± 0.1) of liver hepcidin expression compared with control rats pair-fed with the regular Lieber De Carli diet (1 ± 0) [Fig. 1A (cont.), (alc.)]. Moreover, the iron-induced up-regulation of hepcidin mRNA expression (4.76 ± 0.1) was also suppressed by alcohol exposure (1.7 ± 0.5), compared with control rats (1 ± 0) fed with regular diet [Fig. 1A, (iron), (iron & alc.)]. The median response differences in hepcidin mRNA expression between alcohol, iron, and both iron and alcohol-treated rats were statistically significant (P < 0.0001). The rats fed the ethanol-supplemented Lieber De Carli liquid diet displayed a 2-fold increase in liver iron content (219.5 ± 118 mg/ kg) compared with control rats pair-fed with the regular Lieber De Carli diet (91.5 ± 27 mg/ kg), which was not statistically significant (P = 0.07). The iron content of the livers was similar in rats treated with iron alone (2,093 ± 573 mg/ kg) and in rats treated with both iron and ethanol (2,030 ± 246 mg/ kg). However, the expression of iron storage protein, ferritin in the liver, was significantly (P = 0.02) increased in rats treated with alcohol (3.11 ± 0.98) compared with control rats fed with regular diet (1 ± 0) [Fig. 1B (cont.), (alc.) and C]. The increase in ferritin protein expression was similar in rats treated with iron alone (11.83 ± 0.9) and in rats treated with both iron and alcohol (12.29 ± 1.11) compared with control rats fed with regular diet (1 ± 0) [Fig. 1B (iron), (iron & alc.) and C].

Figure 1.

The combined effect of iron and ethanol on liver hepcidin and ferritin expression in rats fed with Lieber De Carli liquid diets. (A) Rat hepcidin mRNA expression: Complementary DNA was synthesized from liver RNA of Wistar rats fed with regular (cont.), ethanol (36% of calories as ethanol)-containing (alc.), iron-containing (iron), or both iron-containing and ethanol-containing (iron & alc.) Lieber-De Carli liquid diets for 7 weeks, as described in Materials and Methods. Complementary DNA was employed in real-time PCR to detect hepcidin mRNA expression. Hepcidin expression in treated rats was expressed as -fold of that in pair-fed control rats fed with regular diet (cont.). The median response differences in hepcidin expression between alcohol-treated, iron-treated, and both iron-treated and alcohol-treated rats were statistically significant (P < 0.0001). (B) The expression of ferritin heavy chain (H-ferritin) and glyceraldehyde-3-phosphate dehydrogenase (gapdh, as control to confirm equal protein loading) proteins in liver lysates of Wistar rats fed with regular (cont.), ethanol-containing (alc.), iron-containing (iron), or both iron-containing and ethanol-containing (iron & alc.) Lieber-De Carli liquid diets for 7 weeks was determined by western blotting, as described in Materials and Methods. (C) Autoradiographs were scanned by a densitometer and liver H-Ferritin protein expression in each sample was quantified by normalizing to gapdh protein expression. H-ferritin protein expression in the liver of Wistar rats fed with ethanol-containing (alc.), iron-containing (iron), or both iron-containing and ethanol-containing (iron & alc.) liquid diets was expressed as fold expression of that in rats fed with regular (cont.) liquid diet.

In agreement with our previously published results, mice fed only 10% ethanol alone exhibited a significant decrease in hepcidin mRNA expression [Fig. 2A (cont.), (alc.)]. Mice fed with 2% carbonyl iron alone for 3 weeks displayed a 3-fold (3.022 ± 0.637) up-regulation of hepcidin mRNA expression compared with the control mice (1 ± 0) [Fig. 2A, (cont.), (iron)]. Similar to the rat model (Fig. 1), mice treated with iron and alcohol displayed a 3-fold down-regulation of hepcidin expression (1.11 ± 0.03) compared with the mice fed with iron alone (3.022 ± 0.637) [Fig. 2A, (iron), (iron & alc.)]. The median response differences in hepcidin mRNA expression between alcohol, iron, and both iron- and alcohol-treated mice were statistically significant (P = 0.0003). The liver iron was visualized by Prussian Blue staining. No iron staining was observed in alcohol-treated mice, and mice treated with iron and alcohol exhibited similar liver iron content compared with mice treated with iron alone (data not shown).

Figure 2.

Liver hepcidin and duodenal ferroportin expression in mice treated with iron and ethanol. (A) Mouse hepcidin 1 expression: Liver RNA isolated from untreated (control) mice (cont.), mice exposed to 10% ethanol (alc.) for 1 week only, or fed with the iron-overload diet for 3 weeks (iron), or exposed to 10% ethanol for 1 week subsequent to feeding with the iron diet for 3 weeks (iron & alc.), as described in Materials and Methods, was employed to synthesize complementary DNA. Five mice were employed in each group for every experiment (n = 2). Hepcidin1 mRNA expression in treated mice, detected with real-time PCR, was expressed as fold hepcidin expression of that in untreated (control) mice (cont.). The median response differences in hepcidin expression between alcohol-treated, iron-treated, and both iron-treated and alcohol-treated mice were statistically significant (P = 0.0003). (B) Anti-rabbit ferroportin antibody characterization: Wild-type 129/Sv mouse liver lysate proteins resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis were transferred to a polyvinylidene fluoride membrane. The membrane was blotted with similar dilutions of pre-immune rabbit serum, as control (Pre-imm.), anti-ferroportin rabbit serum (FPN), or anti-ferroportin rabbit serum mixed with 0.2 mM ferroportin peptides (see Materials and Methods) to compete for antibody-binding sites (FPN + Pept.). (C) Duodenal ferroportin protein expression: The expression of ferroportin and glyceraldehyde-3-phosphate dehydrogenase (gapdh, as control) proteins in duodenal lysates of untreated (control) mice (cont.), mice exposed to 10% ethanol (alc.) for 1 week only, or fed with the iron-overload diet for 3 weeks (iron), or exposed to 10% ethanol for 1 week subsequent to feeding with the iron diet for 3 weeks (iron & alc.) were determined by western blotting, as described in Materials and Methods. (D) Autoradiographs were scanned by a densitometer, and duodenal ferroportin protein expression in each sample was quantified by normalizing to gapdh protein expression. Ferroportin protein expression in the duodenum of mice treated with alcohol (alc.), iron, and both iron and alcohol (iron & alc.) was expressed as fold expression of that in untreated (control) mice (cont.).

We have previously shown that alcohol elevates the expression of iron transporters divalent metal transporter 1 and ferroportin in the duodenum of both 129/Sv and C57BL/6 mice.3 Hence, the combined effect of iron and alcohol on the duodenal expression of ferroportin protein was investigated. For these studies, we employed an anti-rabbit ferroportin antibody, as described in Methods. The characterization and the specificity of this antibody is shown in Fig. 2B. In agreement with our previously published results, the expression of ferroportin in the duodenum of alcohol-treated mice was elevated (3.8 ±1) compared with untreated (control) mice (1 ± 0) [Fig. 2C (untr.), (alc.) and D]. The expression of ferroportin protein in iron-treated mice was negligible (1.1 ± 0.3) compared with control mice (1 ± 0) [Fig. 2C (untr.), (iron) and D]. Interestingly, the duodenal expression of ferroportin was significantly (P = 0.0037) increased in mice treated with both iron and alcohol (2.8 ± 0.5), compared with mice treated with iron alone (1.1 ± 0.3) [Fig. 2C (iron), (iron & alc.) and D].

To investigate the effect of alcohol on genetic iron overload–mediated hepcidin mRNA expression, we treated Hfe-deficient (Hfe−/−) mice, an animal model for genetic hemochromatosis, with 10% ethanol in the drinking water for 7 days. In accordance with previous reports in the literature, untreated Hfe−/− mice displayed a 2-fold decrease in hepcidin mRNA expression (0.51 ± 0.1), compared with untreated 129/Sv mice (1 ± 0) (Fig. 3A, lanes 1, 2). However, Hfe−/− mice exposed to alcohol displayed a more prominent decrease (0.24 ± 0.06) in liver hepcidin mRNA expression compared with untreated 129/Sv mice (1 ± 0) (Fig. 3A, lanes 1, 3). Alcohol-treated Hfe−/− mice displayed a significant (P < 0.0001) 2-fold decrease in hepcidin expression compared with untreated Hfe−/− mice (Fig. 3A, lanes 2, 3). The alcohol-induced decrease in hepcidin mRNA expression in Hfe−/− mice was also confirmed by northern blotting (Fig. 3B, compare lane 1 and lanes 2–4).

Figure 3.

The effect of alcohol on hepcidin expression in Hfe-deficient (Hfe−/−) mice. (A) Complementary DNA was synthesized from liver RNA of untreated 129/Sv mice (lane 1), untreated Hfe−/− mice (lane 2), or Hfe−/− mice treated with 10% ethanol for 1 week (lane 3) was employed as a template to detect hepcidin1 mRNA expression by real-time PCR. Hepcidin expression in Hfe−/− mice was expressed as fold hepcidin expression of that in untreated 129/Sv mice. (B) Northern blot analysis of mouse hepcidin and mouse beta-actin (as control) mRNA expression in livers of untreated Hfe−/− mice (lane 1) or Hfe−/− mice treated with 10% ethanol for 1 week (lanes 2-4) was performed as described in Materials and Methods.

The transcription factor C/EBPα alpha plays a role in the regulation of hepcidin gene expression.3, 15 We therefore investigated the combined effect(s) of alcohol and iron on C/EBPα alpha DNA-binding activity by gelshift assays, as described in Materials and Methods. As an internal control for the DNA-binding activity of C/EBPα, nuclear lysates from HEK293 cells overexpressing the 42-kDa C/EBP alpha isoform were employed, as described in Materials and Methods [Fig. 4A, (HEK)]. Slightly different electrophoretic mobility was observed with HEK293 nuclear lysates compared with those from mouse livers [Fig. 4A, compare (HEK) with (cont.)]. In HEK293 cells, the transfected C/EBPα forms homodimers of the 42-kDa C/EBP alpha isoform. The liver expresses both 42-kDa and 30-kDa C/EBPα isoforms, as well as C/EBPβ protein, which form heterodimers and migrate faster (N. Timchenko, unpublished data). In agreement with our previously published results,3 the DNA-binding activity of C/EBPα was inhibited in alcohol-treated mice compared with untreated (control) mice [Fig. 4A, (cont.), (alc.)]. C/EBP activity was increased in iron-treated mice compared with untreated mice [Fig. 4A, (cont.), (iron)]. Interestingly, C/EBP DNA-binding activity was also diminished in mice treated with both iron and alcohol compared with mice treated with iron alone [Fig. 4A, (iron), (iron & alc.)]. The addition of antibodies to C/EBPα in the binding reaction showed that the DNA-binding activity of C/EBPα represented most of the activity (Fig. 4A). Quantitative analysis of total C/EBP DNA binding activity confirmed these results (Fig. 4B). Similar results were obtained with each gelshift assay except for slight variations in mice treated with iron (data not shown).

Figure 4.

The combined effect of iron and alcohol on C/EBP DNA-binding activity. (A) C/EBP DNA-binding activity was determined by electrophoretic mobility gelshift assay, as described in Methods. Two micrograms nuclear lysate protein isolated from HEK293 cells overexpressing C/EBP alpha (HEK) was employed as an internal control for the position of homodimers of the 42-kDa isoform of C/EBPα. Five micrograms nuclear lysate protein isolated from the livers of untreated (control) mice (cont.), mice exposed to 10% ethanol for 7 days (alc.), iron-treated for 3 weeks (iron), or exposed to 10% ethanol for 7 days after the iron treatment (iron & alc.) was employed in the gelshift assays. Antibodies to C/EBPα (anti-α Ab) were incorporated into the binding reaction before probe addition to determine the specificity of C/EBPα DNA-binding. (B) Total C/EBP DNA-binding activity (without C/EBPα antibody incubation) in liver lysates was quantified by scanning autoradiographs using a densitometer. DNA-binding activity in liver lysates of mice treated with alcohol (alc.), iron, or both iron and alcohol (iron & alc.) was expressed relative to C/EBP activity in untreated (control) mice (cont.).

Discussion

Iron is considered to be one of the secondary risk factors in alcoholic liver disease.16 Both iron and alcohol can individually cause oxidative stress and lipid peroxidation.16, 17 Hence, understanding the molecular mechanisms underlying the interaction of alcohol and iron in the liver is important. Hepcidin, synthesized by the liver, plays a central role in iron homeostasis and protects the body from the harmful effects of iron by inhibiting iron uptake and iron release.2 The purpose of this study was to define the combined effect of iron and alcohol in the regulation of hepcidin mRNA expression in vivo.

Recently, we have demonstrated that both 129/Sv and C57BL/6 mice exposed to ethanol in the drinking water for 7 days display significant down-regulation of liver hepcidin mRNA expression.3 Bridle et al.4 have reported down-regulation of hepcidin mRNA expression in rats fed with ethanol-Lieber De Carli diet for 12 weeks. In this study, we compared the combined effect of alcohol and iron on hepcidin mRNA expression in the mouse and rat models.

Iron overload increases liver hepcidin synthesis2 and, accordingly, we have observed elevated hepcidin mRNA expression in mice and rats treated with iron. In accordance with our previously published results and the results in the literature, alcohol reduced hepcidin mRNA expression in both the mouse and the rat model of alcohol exposure.3, 4 However, the down-regulation of hepcidin expression in our rat model was not as prominent as reported by Bridle et al.4 This may be because the duration of ethanol exposure was 5 weeks shorter in our rat model. However, we observed an increase in the liver-specific expression of the iron storage protein, ferritin, in our rat model.

Alcohol abolished the up-regulation of hepcidin mRNA expression by iron, yielding hepcidin expression levels similar to those in the untreated animals. Both acute (mice exposed to 10% ethanol in drinking water for 7 days) and chronic (rats pair-fed with ethanol-Lieber De Carli diet for 7 weeks) alcohol exposure was capable of negating the effect of iron on liver hepcidin mRNA expression. Moreover, the suppression of iron-mediated up-regulation of hepcidin mRNA expression by alcohol occurred irrespective of whether alcohol was introduced simultaneously with iron (rat model) or subsequent to iron overload (mouse model). These results suggest that alcohol nullifies the effect of iron on hepcidin expression, independent of the duration of ethanol exposure or the state of body iron stores.

The inhibition of hepcidin synthesis by alcohol should lead to a further increase in iron uptake and storage. Accordingly, we observed an increase in the liver-specific expression of the iron storage protein ferritin in rats maintained for 7 weeks on the ethanol–Lieber De Carli diet compared with pair-fed control rats. Tsukamoto et al.8 also reported an increase in ferritin expression in liver macrophages isolated from rats that which had received a continuous intragastric infusion of ethanol for 9 weeks. However, we did not observe a further increase in the iron content or ferritin protein expression in the liver of rats treated with both iron and ethanol, compared with rats treated with iron alone. This may be because the liver is already saturated with iron in this experimental model. The liver iron content of 4- to 20-week-old rats fed a standard commercial diet has been reported to be between 60 and 115 mg/ kg liver.18 We observed at least 10-fold higher iron content in the livers of rats treated with iron or both iron and alcohol.

We have previously reported an increase in the duodenal expression of ferroportin protein in both 129/Sv and C57BL/6 mice treated with ethanol in the drinking water for 7 days.3 This finding was abolished by injecting mice with the hepcidin peptide.3 In the current study, we have confirmed these results. Interestingly, mice treated with both alcohol and iron, but not mice treated with iron alone, also displayed an increase in duodenal ferroportin expression compared with untreated mice. These results suggest that the alcohol-mediated suppression of hepcidin mRNA expression in iron-overloaded mice leads to increased duodenal ferroportin expression. Of note, hepcidin inhibits iron uptake and release by binding to and degrading ferroportin protein.19 Hence, the suppression of iron-induced hepcidin expression by alcohol may lead to increased iron transport, despite the iron overload state in these mice.

Alcohol has been reported to increase the severity of disease in genetic hemochromatosis patients.7 Genetic hemochromatosis patients and Hfe-deficient mice display reduced hepcidin expression.5, 6 Our results demonstrated that alcohol treatment causes a further decrease in liver hepcidin mRNA expression in Hfe−/− mice. Flanagan et al.,20 however, have reported no change in hepcidin mRNA expression in Hfe knockout mice treated with 20% alcohol for 2 weeks. The reasons for this discrepancy are unclear and may be due to differences in experimental design including the generation and genetic background of Hfe−/− mice.10, 20 Our real-time PCR results demonstrating the alcohol-mediated decrease in hepcidin mRNA expression in Hfe−/− mice were also confirmed with northern blotting. The authors suggest that alcohol reduces hepcidin mRNA expression and increases serum iron levels only in 129/Sv mice.20 We have previously reported that both 129/Sv and C57BL/6 mouse strains display reduced liver hepcidin expression3 (see Supplementary Fig. 1). Moreover, the down-regulation of hepcidin mRNA expression led to increased duodenal divalent metal transporter 1 and ferroportin expression in both strains of mice.3 Our findings, therefore, suggest that alcohol abolishes the effect of both dietary and genetic iron overload on hepcidin mRNA expression. The further decrease in hepcidin expression in alcohol-treated Hfe−/− mice suggests that this may be one of the mechanisms by which alcohol increases the severity of disease in genetic hemochromatosis patients.

To define the mechanisms by which alcohol abolishes iron-mediated hepcidin transcription, we have investigated the role of the transcription factor, C/EBPα. C/EBPα protein expression is elevated by iron overload.15 The transcription factor C/EBPα plays a role in the regulation of hepcidin gene expression.3, 15 We have shown that the inhibition of C/EBPα activity by alcohol is involved in the down-regulation of hepcidin 1 gene transcripts in the liver.3 Reduced C/EBPα mRNA expression was reported in rats with chronic ethanol exposure.4 In agreement with our previously published results, mice treated with alcohol displayed reduced C/EBP activity.3 C/EBP DNA-binding activity was increased by iron overload. Interestingly, treatment with alcohol abolished the iron-induced up-regulation of C/EBP DNA-binding activity. Thus, these results suggest that the inhibition of C/EBP DNA-binding activity by alcohol may play a role in the suppression of iron-induced hepcidin transcription in animals treated with both alcohol and iron. We have previously demonstrated that alcohol-induced oxidative stress reduces C/EBP alpha protein expression and activity in vivo, which was reversed by treating mice with antioxidants.3 It is therefore plausible that alcohol-mediated oxidative stress may play a role in the suppression of iron-induced hepcidin transcription by alcohol.

Collectively, our data demonstrate that alcohol suppresses the iron-induced up-regulation of liver hepcidin transcription by inhibiting the activation of C/EBPα. Our findings also suggest that the mechanisms that protect the body from the harmful effects of iron overload (for example, increased hepcidin expression and decreased iron uptake and storage) are compromised by alcohol. Further research is required to determine whether the effect of alcohol on iron-regulatory proteins plays a role in the progression of liver injury observed in alcoholic liver disease and in genetic hemochromatosis patients in the presence of excessive alcohol intake.

Acknowledgements

The authors thank Elizabeth Lyden for assistance with statistical analysis.

Ancillary

Advertisement